Benchmarking WM

P a g e | 1 P a g e | 1
Benchmarking Document
Technology: Washing machines (clothes washers)
The information and analysis contained within this summary document is developed to inform policy makers. Whilst the information analysed was supplied by
representatives of National Governments, a number of assumptions, simplifications and transformations have been made in order to present information that is
easily understood by policy makers, and to enable comparisons with other countries. Therefore, information should only be used as guidance in general policy - it
may not be sufficiently detailed or robust for use in setting specific performance requirements. Details of information sources and assumptions, simplifications and
transformations are contained within the document or the related Mapping Documents.

Issue date: April 2012
Other regions covered:
China, EU
Other funding countries:
France, Netherlands,
Japan, South Africa,
Sweden

Participating countries:
Australia, Austria, Canada,
Denmark, Republic of Korea,
Switzerland, UK, USA

Benchmarking report for
Washing Machines (clothes washers)
Issue Date: 28 April 2012 (Danish data updated J uly 2012)

For further information refer to http://mappingandbenchmarking.iea-4e.org/matrix
or email operating.agent@mapping.iea-4e.org
`
P a g e | 2 P a g e | 2
Benchmarking Document Washing machines



Summary for policy makers
This benchmarking has been undertaken as part of the IEAs Mapping and Benchmarking
Annex of the Efficient End-Use Electrical Equipment (4E) Implementing Agreement.
The analysis has been undertaken to provide policy makers with comparisons of the
performance of new washing machines (clothes washers) over time and is designed to
provide policy makers with a broad analysis of:
Key differences in product performance between countries;
Major outcomes of policy interventions to date;
Areas of concern for policy makers, including areas where policy intervention may be
desirable in the future.

Due to differences in the test methodologies used in these countries/regions, adjustment or
normalisation of the original datasets is required to present consistent and comparable
information to enable benchmarking of product performance between the different
jurisdictions. However, when applied to washing machines, such an approach leads to some
inherent issues of which readers should be aware. In particular there is a strong
interrelationship in performance variables, i.e. the energy performance of washing machines
is affected by a range of interdependent variables including washing temperature; washing
cycle time; wash quality; spin effectiveness; rinse effectiveness; the type and size of the
laundry load, and a number of other variables. While recognising that these key performance
characteristics are all intrinsically linked, no public domain information has been identified
that has allowed the Mapping and Benchmarking Annex to compare washing machines of
differing performance across all these variable types. Therefore readers should be aware
that:
Normalisation has been based on the energy used for water heating only which has
limitations both in approach and due to the optimisation of machines to local
requirements. In particular the approach is highly sensitive to variations in wash test
temperature and water consumption.
Individual performance attributes of the washing machines (wash quality, spin
effectiveness, etc) have not been normalised, nor have they been accounted for in
the normalisation of energy consumption.
The use of the washing machine by consumers may be significantly different from the
conditions under test. The data and analysis on energy use and other performance
attributes presented in this report are based on testing outcomes and are likely to be
different from those experienced by the consumer.
For these reasons, the resulting comparison of washing machine performance between
jurisdictions has significant levels of uncertainty. Additional limitations in the normalisation
process have a particular impact on the data presented for Canada and the USA. Therefore,
normalised Canadian and USA data results are less comparable than normalised results
`
P a g e | 3 P a g e | 3



from other countries, although the relative positioning of results from Canada and the USA
are highly likely to be comparable.

Due to these significant uncertainties, and the need for reliable evidence when drawing
conclusions for use by policy makers, it has not been possible to draw definitive conclusions
on the potential for improvement in product efficiencies between countries. However, the
following information drawn from the analysis may be of value to policy makers:
The combination of frequently revised MEPS and labelling has had a very strong
impact in Canada and the USA (although the biggest market response seems to
align primarily with the introduction/revision of MEPS). Ultimately Canadian and US
unit energy consumptions are now (likely) to be broadly in line with most European
countries despite being at significantly higher initial levels. Therefore it is reasonable
to conclude that challenging and regularly revised MEPS are a highly effective
method of reducing consumption.
The impact of labelling within the EU countries, and the associated voluntary
agreement with industry, has had a mixed effect. Across the EU as a whole and
within Austria, there has been a significant market response. However, within the UK
improvement has been minimal. Therefore policy makers should be aware of the
potential improvements that can be achieved through the combined application of
labelling and voluntary agreements. However, ongoing monitoring of the market
should be undertaken to ensure policy impacts are occurring in all market segments
and, where this is not the case, revisions to the agreement/policy should be
considered.
In the one country reporting which relies solely on labelling as its policy intervention
(Australia), improvements have been seen in overall average unit energy
consumption of washing machines, but the performance of these units still lags
significantly behind all other participating countries
1
.
The implementation of policy to drive improvement in the energy performance of top
loading washing machines, or to encourage consumer switching from top loader
machines to their front loader competitors, would yield significant energy savings in
those countries where top loader penetration is still high. However, policy makers
should be aware that such action may adversely affect other performance variables,
in particular wash cycle times.
The rated load capacities of washing machines (i.e. the amount of laundry that a unit
can hold in a single wash cycle) are increasing in almost all jurisdictions and there is
no indication that these increases are reaching a plateau in any country. This
ongoing increase in volume is at least partly responsible for increasing product
efficiency (more so in recent years where the improvement in unit energy

1
However, the significant use of cold water in washes by Australian consumers (which is not reflected in the test
methodology) may negate the need for policy intervention to address this apparent difference.
`
P a g e | 4 P a g e | 4



consumption is beginning to stagnate in some regions). Hence policy makers may
wish to investigate whether the actual size of loads washed by consumers is
increasing in line with increasing machine sizes. If actual load sizes are not
increasing significantly, it is likely that further improvements in declared product
efficiencies due to increasing product sizes may be disguising stagnation in actual
energy consumption per unit of clean laundry. If this is the case, consideration should
be given to limiting unit size and/or capping energy consumption directly (or a
combination of both through amendments to algorithms that define minimum
efficiency levels).
The load capacities of top loading washing machines were traditionally greater than
front loading units. However, this trend has reversed, with average front loader
capacities now larger than average top loader capacities for all markets where data
are available.
Total water consumption per cycle has been reducing over a number of years but
has recently reached a plateau in many countries (at significantly differing levels).
Hence, it is possible that a point is approaching where reductions in water
consumption can no longer be sustained without significant deterioration in wash
performance and/or rinse effectiveness, or increased cycle time. Therefore, in
countries where water and/or energy consumption/efficiency will continue to be a
focus of policy intervention, policy makers should consider increasing vigilance
regarding wash and rinse performance to ensure their policy intervention will not
result in impaired unit performance to which consumer reaction may be negative. .
Spin effectiveness is improving in all countries where it is measured. However, as
manufacturers strive to reduce energy consumption, it is possible that spin
effectiveness may be reduced. Therefore, policymakers may wish to keep a watching
brief on spin efficiency to ensure consumers remain satisfied with spin performance,
and to ensure that improvements in washing machine energy consumption through
reduced spinning are not resulting in significantly greater increases in consumption in
the post-wash drying of the laundry.

`
P a g e | 5 P a g e | 5



Contents

Summary for policy makers 2
1 Introduction 6
1.1 Important cautions for interpreting and using mapping and benchmarking information 7
2 Comparison of energy consumption and efficiency 12
2.1 Observations (excluding Canada and the USA) 12
2.2 Observations (including the USA and Canada) 22
2.3 Key issues for policy makers 27
3 Non energy performance variables 29
3.1 Observations 29

Annexes
Annex 1 Terminology used 33
Annex 2 Framework for grading mapping and benchmarking outputs 35
Grading of data/mapping outputs 35
Grading of comparison between country outputs (benchmarking) 36
Annex 3 Categorisation of original datasets, approach to normalisation and
associated cautions 37
Original data quality 37
Benchmarking information 37
Important cautions 44
Summary grading of mapping and benchmarking data 45
Annex 4 Policy summary table 49

`
P a g e | 6 P a g e | 6



1 Introduction
This benchmarking has been undertaken as part of the IEAs Mapping and Benchmarking
Annex of the Efficient End-Use Electrical Equipment (4E) Implementing Agreement.

The analysis seeks to provide policy makers with comparisons of the performance of new
and installed washing machines (clothes washers) over a period of years and is designed
to provide policy makers with a broad analysis of:
Key differences in product performance between countries;
Major outcomes of policy interventions to date;
Areas of potential concern, including areas where policy intervention may be
desirable in the future.
This benchmarking addresses all washing machines excluding:
Twin-tub units;
Washer/dryers (single and two drum);
Machines with capacity less than 1 kg or bigger than 13 kg
2
.
Data was sought from all countries participating in the Annex and a number of additional
countries/geographical areas. The countries/areas for which information was available and
could be sourced were:

Australia, Austria, Canada, China, Denmark, the Republic of Korea,
Switzerland, the UK, the USA. and a sample of countries that are representative
of the EU.

Information was sought for the period 1996 to 2009 relating to product energy efficiency,
product energy consumption, water consumption, product capacity, etc. For individual
countries and regions, information obtained was mapped in a consistent format and
presented to show:
The energy efficiency and energy consumption of new products sold within individual
markets;
The overall changes in new products for functional performance that affect energy
consumption including load capacity, water consumption, wash quality, spin
efficiency, rinse efficiency, etc;
Policies that are thought to have influenced the performance of new products and
stock;
Changes in products within the stock (products in use in households) over the period;
Cultural issues that may have influenced product selection within individual countries.

2
Please refer to the Mapping and Benchmarking Annex website to review the detailed product definition, data
request and individual country mappings at:
http://mappingandbenchmarking.iea-4e.org/matrix?type=product&id=3
`
P a g e | 7 P a g e | 7



The comparisons of the reported product performance presented in the individual mappings
form the basis of this analysis.
1.1 Important cautions for interpreting and using mapping and
benchmarking information
1.1.1 Original data quality
Significant efforts have been made by all participating parties to obtain and process data
from a range of sources and to ensure the integrity of the data supplied. However, inevitably
there have been some difficulties sourcing information for all countries/regions, and indeed
in sourcing all information from individual countries/regions even where this information
exists. Therefore, the specific nature of each dataset is different. For example, some
datasets are based on detailed information on individual models across a whole market;
others are based on averages of aggregated data from the whole market, and some
datasets are selections/samples which may be representative of the market as a whole or
just a subset of the market (for example the best performing products).

Further, in some cases, data adjustment is required to make the material comparable
between countries (for example, the conversion of North American unit capacities defined by
the overall drum size to the equivalent load capacity in kilogrammes used elsewhere).

Thus, to provide readers with an indication of the relative reliability of a particular dataset
within the context of the other data being presented, the Mapping and Benchmarking Annex
has developed the Framework for Grading Mapping and Benchmarking Outputs
3.
This
framework enables the allocation of gradings based on a robust, indicative and illustrative
scale. The original data received (including any manipulations necessary
4
) have been
classified based on this framework with the associated gradings shown in Figure 1 (refer to
Annex 3 for a summary of the justification for individual data classifications and significant
associated cautions).
1.1.2 Benchmarking approach and related quality of outputs and cautions
To enable the comparison of product performance between countries, the Mapping and
Benchmarking Annex compares products based on their performance when undergoing the
standard test defined in the local methodology/regulations. Differences in the local test
methodologies are then normalised in an effort to make each original dataset supplied
comparable with those from elsewhere.

However, when applied to washing machines, such an approach leads to some inherent
issues of which readers should be aware. In particular there is a strong interrelationship in
performance variables, i.e. the energy performance of washing machines is affected by a

3
This Framework is generally used across all Mapping and Benchmarking outputs (refer to Annex 2).
4
All manipulations of individual data sets are detailed in the associated country/region mapping at
`
P a g e | 8 P a g e | 8



range of interdependent variables including washing temperature; washing cycle time; wash
quality; spin effectiveness; rinse effectiveness; the type size, material and number of
garments that make up the load to be washed; the total size of load, and a number of other
variables. Even where the same test method is used, there is a great deal of uncertainty
over how these factors interrelate for an individual machine. This means the development of
normalisation factors related to an individual variable (for example energy consumption) will
have inherent shortcomings as they do not factor in the positive or negative impact on other
variables. For example, units tested at a specific test temperature may provide a particularly
level of washing performance which may vary at an alternative temperature, hence simply
adjusting energy consumption based on the energy required to heat the water to an
alternative wash temperature does not reflect the full impact on the overall machine
performance.

Having recognised that key performance characteristics (energy consumption, load type and
size, wash quality, etc.) are all intrinsically linked:
No public domain information has been identified that has allowed the Mapping and
Benchmarking Annex to compare washing machines of differing performance across
all the performance variables;
There is limited public domain information on the impact of the various testing
methodologies on the reported overall performance of individual units, or indeed, to
convert the individual performance characteristics where they are measured
differently.
Within the context of the specific datasets received, these issues are further compounded by
the provision by some countries of product level data which allows analysis of the
interrelation of many variables for individual machines, and the provision of aggregated
market data from elsewhere which makes such analysis impossible.

Therefore, analysis and reporting within this benchmarking is restricted to:
Normalisation of energy consumption based on a correction for nominal
test/performance standard water inlet temperatures and wash temperatures;
Data that is as declared under local test methodologies for all the individual
performance characteristics of the washing machines other than energy. No
normalisation is undertaken to account for variations in the measurement of
these variables between testing regimes, nor for the associated impact on
energy consumption.
Accurate comparisons are further challenged by the following issues:
Washing machines are optimised to local conditions. Therefore, simple correction of
water temperature is not thoroughly robust as units tested at a lower temperature
(typically) require less water to achieve the same wash performance when operated
`
P a g e | 9 P a g e | 9



at higher wash temperatures
5
. Hence normalisation using the original water quantity
will not be wholly representative of the performance of all units.
There is some empirical evidence
6
to suggest washing machines under test in some
jurisdictions do not actually reach the test temperature specified in the test
methodology and hence normalisation of energy consumption based only upon test
temperatures may not be accurate.
The difference in the requirements of individual testing methodologies and the
normal use of washing machines by consumers adds one further complication to the
interpretation of benchmarked data, i.e., the actual energy consumption of the
washing machines will vary significantly as consumers will often wash at
temperatures very different to those specified in test standards. While this is relevant
to all datasets, it is known to be of particular relevance to benchmarking data for
Australia. Australian normalisation has been undertaken based on the nominal warm
wash temperature defined in the local standard. However consumers are known to
perform a large number of washes in cold water (at the temperature of water intake
or similar), particularly in top loading washing machines. Hence, comparative
benchmarking of Australian data, particularly those data associated with top
loading washing machines, should be treated with some caution
7
.
.

Normalised Canadian and USA data results are less comparable than normalised
results from other countries (although the relative positioning of results from Canada
and the USA are highly likely to be comparable). This is partly due to a shortcoming
of the normalisation methodology which is based on a unit energy consumption value
for a specific set of operating conditions, where as the original average unit energy
consumption declared in Canada and the USA is based over a range of cycle
conditions (in particular the quantity and temperature of water used). This
shortcoming in the normalisation methodology is compounded by a revision to the
Canadian and USA required test method/energy reporting algorithm in 2004. This
revision included changes to the balance of wash temperatures, such that the hot
and warm wash cycles were assigned lower weighting factors after 2004. This
revision, on average, will tend to result in higher energy consumption values being
reported prior to 2004 in comparison with those after
8
. Further, the load (tub)
capacities of washing machines in Canada and the USA are declared based on the
physical internal dimensions of the machines which require conversion to kilogramme
loads to be comparable with data reported from elsewhere. Thus, benchmarked
data from Canada and the USA is presented in separate sections. The results
presented in these sections should be interpreted in the context of the

5
The cleaning performance of a washing machine is governed by a combination of detergent, temperature,
quantity of water, the degree of agitation the laundry experiences and the cycle time. If all things remain equal
and the water temperature is increased, less water will be required to achieve the same wash performance.
6
Unpublished results from tests within the EU indicate that units rarely reach the specified nominal 60
C

test
temperature during washes, and the average temperature is significantly below the nominal test. This situation
may or may not occur in other jurisdictions.
7
Refer to Annex 3, in particular the section Test Declarations, Actual Water Temperature and Normal Usage for
details of the sensitivity of the normalisation approach and benchmarking outcomes to temperature variation.
8
For more detailed information on the revision and associated impact, please refer to the USA mapping
document at http://mappingandbenchmarking.iea-4e.org/matrix?type=product&id=3.
`
P a g e | 10 P a g e | 10



limitations in comparability with washing machine performance from
elsewhere as outlined above.
1.1.3 Important cautions
The above descriptions of the limitations of original data and the normalisation approach
employed for the cross-country/region comparisons lead to the following summary cautions
of which readers should be aware:
All original data and the benchmarking results derived from them are not 100%
comparable and have been graded to provide an indication of the
quality/comparability.
Normalisation has been based on the energy used for water heating only which
has limitations as an approach as it is highly sensitive to variations in wash test
temperature and water consumption, and does not account for optimisation of
machines to local requirements.
Individual performance attributes of the washing machines (wash quality, spin
effectiveness, etc) have not been normalised, nor have they been accounted for in
the normalisation of energy consumption.
conditions under test. The data analysis on energy use and other performance
attributes presented in this report are based on reported testing outcomes which are
likely to be different from those experienced by the consumer.
The original data from Canada and the USA is in a form that requires significant
manipulation to be comparable with data supplied from elsewhere. The level of
manipulation required is substantial and introduces uncertainty in the level of direct
comparability with other countries, although the relative positioning of results from
Canada and the USA is highly likely to be comparable. Therefore, benchmarked data
from the Canada and the USA is presented alongside data from elsewhere, but in a
separate section and should be interpreted in the context of the limitations in
comparability.
For a fuller description of original data quality, the approach to normalisation of data to
enable comparison benchmarking between countries, and the associated cautions, please
refer to Annex 3. However, Figure 1 provides a summary of gradings of both original
mapping data and the associated benchmarking outputs.

`
P a g e | 11 P a g e | 11



Figure 1. Summary classification of original data and benchmarking outputs.
Country Data classification and limitations
Australia Sales weighted source data is Robust, product weighted source data is Indicative.
Benchmarking outputs: Sales weighted information is Indicative, product weighted
information is Illustrative. However, extreme caution should be used when
comparing with other countries given the known consumer preference for cold wash
(resulting in significantly lower energy consumption that that shown).
Austria All source data is Indicative.
Benchmarking outputs: All information is Illustrative.
Canada All source data is Indicative.
Benchmarking outputs: All information is Illustrative and, due to known limitations in
the benchmarking approach, the degree of comparability with other countries is
limited.
China All source data is Illustrative.
Benchmarking outputs: Due to the uncertain nature of the quality of this data
source, information on China is excluded from the analysis.
Denmark All source data is Indicative.
Korea All source data is Indicative.
Benchmarking outputs: Data is Illustrative (front-loaders only). Due to the lack of
definition of a test temperature for top-loading units, these units have been excluded
from the benchmarking analysis.
Switzerland Sales weighted source data is Indicative:
Benchmarking outputs: Data is Illustrative
UK Front-loader sales weighted source data is Robust, front-loader weighted source
data is Indicative.
information is Illustrative. (Note: all full market data is based solely on front-loader
data given the extremely limited nature of top-loader penetration in the market).
USA Sales weighted source material is Indicative.
Benchmarking outputs: All information is Illustrative and, due to known limitations in
the benchmarking approach, the degree of comparability with other countries is
limited.
EU All source data is Indicative.

Given the significant differences in measurement methodologies and reporting
protocols, all non-energy variables, i.e. load, water consumption, drying effectiveness
and wash quality, are presented as illustrative.
`
P a g e | 12 P a g e | 12



2 Comparison of energy consumption and efficiency
2.1 Observations (excluding Canada and the USA)
2.1.1 Unit energy consumption
With the exception of Switzerland and recently Denmark, all countries have reported a fall in
the sales weighted (manufacturer declared) unit energy consumption (UEC) of new washing
machines when tested under local test conditions over the reporting period (refer to Figure 2).
In some cases, this fall has been substantial, with UEC in Australia falling by 26% over the
2001 to 2007 period
9
, although from a relatively high starting point. The primary policy in
force in Australia was mandatory labelling and product registration from 1998 onwards
10
.

The fall in UEC within the EU countries as a whole appears to still be significant at 19% over
the slightly longer 2000 to 2007 period, with this fall mirrored in the individual case of Austria
(16%). However, falls in the UEC in UK washing machines were well below the EU average
at 6%
11
. This is despite mandatory EU labelling across all EU countries from 1996, and two

9
Reported UEC in Australian prior to 2001 is based on a smaller proportion of the market than 2001 and later,
hence the comparison of 2001 with later years. Refer to the Australian Mapping document for more details.
10
The introduction of this requirement, and the associated stricter reporting regime, is thought to account for the
sudden rise in average product UEC shown over the 1998 to 2000 period.
11
The UK UEC in 2000 (1.15 kWh/cycle) was only slightly better than in the EU as a whole (1.17 kWh/cycle). By
2009, the UK UEC lagged most EU reporting countries by at least 10%.
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (40 degC) 1.57 1.55 1.53 1.62 1.73 1.58 1.58 1.34 1.30 1.33 1.26 1.18
Austria (60 degC) 1.18 1.16 1.11 1.06 1.03 1.00 0.99 0.99 1.00 0.99
Denmark (60 degC) 0.77 0.71 0.78 0.73
Switzerland (60 degC) 1.05 1.03 1.04 1.06 1.09 1.06
United Kingdom (60 degC) 1.16 1.15 1.12 1.09 1.06 1.06 1.03 1.06 1.08
EU (60 degC) 1.17 1.10 1.05 1.03 1.00 1.00 0.99 0.95 0.95 0.97
0.0
0.5
1.0
1.5
2.0
S
a
l
e
s

w
e
i
g
h
t
e
d

c
o
n
s
u
m
p
t
i
o
n

(
d
e
c
l
a
r
e
d

k
W
h

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
Solid line = robust data Dashed line = indicative data Dotted line = illustrative data
Figure 2. Sales weighted energy consumption as declared under local test conditions (kWh).
`
P a g e | 13 P a g e | 13



voluntary agreements between the EU and the washing machine producers covering the
period 1996 to 2008
12
. Note that there is a sharp improvement in UEC in the EU as a whole
and within Austria between 2000 and 2003 which may indicate that rapid improvements in
product consumption occurred in EU countries prior to 2000. Such a conclusion is somewhat
supported by the reductions in UEC experienced by Denmark over the 1996-2001 period.
However, given the lack of available data, such a conclusion is rather speculative.

Interestingly, in Switzerland, where use of the EU label became mandatory in 2003 and
MEPS (equivalent to the EU labelling A level) were introduced in 2010, the UEC has
remained broadly fixed, starting at 1.05 kWh/cycle in 2003 and reaching 1.06 kWh/cycle in
2008 (in line with the UK).

However, these apparently significant differences in initial unit energy consumptions and
subsequent reductions in consumption are potentially misleading because of:
Variations in the required testing temperatures;
Potential variation in the relative percentages of top and front-loader units in the
different markets and the associated differences in performance;
Potential variations in the size of load serviced by machines in individual markets.
The impact of each of these issues is examined below.
2.1.2 Normalised unit energy consumption
Figure 3 provides exactly the same unit energy consumption data as Figure 2 but this time
normalised to account for the differences in energy required to heat water for the varying
local test/regulatory requirements (see Annex 3 for details of the normalisation undertaken).

As can be seen, in almost all countries the normalisation process results in the UEC sharply
reducing to between 40-50% of the initial unadjusted levels, the exception being Australia
where UEC rises by approximately 10%. Hence the relative UECs have moved significantly,
with the 2007 Australian UEC now almost 2.5 times higher than that the EU average
compared with just 25% higher prior to normalisation.

However, overall the picture remains broadly similar with approximately equal improvements
being experienced by countries before and after normalisation and average UEC for
products in most countries bunching around the 0.5 kWh/cycle mark in 2007. The exceptions

12
For the EU countries where data are available for 2008-09, reduction in UEC has all but stagnated (and in
some cases increased). It is interesting to note that there is some anecdotal evidence that manufacturers
supplying products to the EU have continued to improve the performance of some units. However, this
improvement in performance would not have been visible to consumers due to the EU labelling regime in place
(as a large majority of units on the market achieve the top A rating and further product improvements would not
be recognised in the label). Therefore, the declarations of unit performance made by manufacturers may have
been under reporting the actual unit performance with a view to rapidly taking advantage of future revisions to the
labelling regime. However, such a conclusion is speculative as the evidence is based on unpublished test data on
a small sample of models over the 2007 to 2009 period, and actual variations in consumption and efficiency are
somewhat masked by changes to unit capacity (see section 2.14).
`
P a g e | 14 P a g e | 14



to this 0.5 kWh/cycle UEC are washing machines in Australia which are using on average
more than twice as much energy per cycle (1.29 kWh/cycle in 2007).
2.1.3 Comparative unit energy consumption based on orientation of the
washing machine
There are significant historical and cultural differences between countries in the orientation
of washing machines used. In general, Australia, Korea and North America consumers have
traditionally favoured top-loading washing machines, and European consumers front-loading
machines. However, more recently there is a migration in all markets towards front-loading
machines, albeit at differing speeds. Therefore, while the preceding comparison gives an
indication of the average UEC of units across the whole market in each country, there is
value in comparing the performance of washing machines by orientation. Such an analysis
gives a more realistic comparison of machines of a similar type and enables the identification
of opportunities that may exist for improvements in each specific machine orientation. Figure
4. and Figure 4 give this normalised Unit Energy Consumption breakdown by the orientation
of the washing machine for countries where this split in data was available.

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (40 degC) 1.78 1.75 1.73 1.84 1.97 1.79 1.79 1.48 1.44 1.48 1.39 1.29
Austria (60 degC) 0.57 0.56 0.53 0.51 0.49 0.47 0.47 0.47 0.47 0.47
Denmark (60 degC) 0.35 0.31 0.35 0.32
EU (60 degC) 0.55 0.52 0.50 0.48 0.47 0.50 0.49 0.48 0.49 0.49
0.0
0.5
1.0
1.5
2.0
S
a
l
e
s

w
e
i
g
h
t
e
d

c
o
n
s
u
m
p
t
i
o
n

(
n
o
r
m
a
l
i
s
e
d

k
W
h

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
Figure 3. Sales weighted normalised energy consumption (kWh).
`
P a g e | 15 P a g e | 15



1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (20+ degC) 1.81 1.81 1.81 1.93 2.10 1.98 1.98 1.79 1.76 1.84 1.82 1.73
Austria (60 degC) 0.53 0.50 0.46 0.47 0.47 0.46 0.46 0.45 0.44 0.44
EU (60 degC) 0.57 0.53 0.50 0.49 0.47 0.47 0.47 0.44 0.44 0.46
0.0
0.5
1.0
1.5
2.0
S
a
l
e
s

w
e
i
g
h
t
e
d

c
o
n
s
u
m
p
t
i
o
n

o
f

t
o
p

l
o
a
d
e
r
s
(
n
o
r
m
a
l
i
s
e
d

k
W
h

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (40 degC) 0.64 0.86 0.65 0.65 0.59 0.59 0.62 0.62 0.66 0.72 0.75 0.77
Austria (60 degC) 0.57 0.57 0.53 0.51 0.49 0.47 0.47 0.47 0.47 0.47
Republic Of Korea 0.62 0.61 0.61
EU (60 degC) 0.55 0.52 0.50 0.49 0.48 0.49 0.49 0.49 0.50 0.51
0.0
0.5
1.0
1.5
2.0
S
a
l
e
s

w
e
i
g
h
t
e
d

c
o
n
s
u
m
p
t
i
o
n

o
f

f
r
o
n
t

l
o
a
d
e
r
s
(
n
o
r
m
a
l
i
s
e
d

k
W
h

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
Figure 5. Sales weighted normalised energy consumption for top-loader machines (kWh).
Figure 4. Sales weighted normalised energy consumption for front-loader machines (kWh).
`
P a g e | 16 P a g e | 16



In the EU as a whole, top-loader UEC has improved by almost 23% compared with 11% for
front-loaders. However, the reverse is true in Austria where front-loader UEC has improved
by almost 18% while top-loaders have improved by 15%. In Australia, this situation is even
more extreme, where top-loader energy consumption has reduced by 15%, but front-loader
consumption has actually increased by 30% (although in Australia, despite these relative
movements, front-loader washing machines still have UECs at half the level of top-loader
units
13
).

These significant changes over the period have resulted in a current situation where:
Average top-loader UECs ultimately become similar in Austria (0.45 kWh) and the
EU as a whole (0.44 kWh). This leaves Australia as the very significant outlier with
top loader UEC of 1.73 kWh, more than three times the EU levels;
Average front-loader UECs have a much smaller variation in 2000 with almost all
countries in the 0.55-0.59 kWh performance range. However, this spread widens by
2007 with Australia increasing to 0.77 kWh, the UK staying broadly flat at 0.52 kWh,
and the EU and Austrian UECs falling to 0.51 kWh and 0.47 kWh respectively.
Thus, the picture to date is rather complex with few clear lessons. However, this mixed
picture of variable UEC may become clearer when the size of the load is taken into account.
By doing so, the energy consumption per unit of washing service can be calculated.
2.1.4 Trends in rated load sizes
The rated size of the load defines the quantity of service that a particular washing machine
can provide, i.e. the quantity of laundry that can be washed in a single cycle. The different
average product weighted rated capacity split between front (single line) and top-loaders
(double line) in each country is shown in Figure 6. As can be seen, almost all countries are
experiencing annual growth in average product volumes
14
. Between 2000 and 2007, where
information is available for the majority of countries, average front-loader rated load capacity
has increased by almost 17%, while top-loader capacity has risen by an average of just 4%.

The lowest growth in top-loaders was experienced by Austria at 2.2%. Loads in Australian
top-loading machines grew considerably, by 8.8%. However, growth in the capacity of front-
loaders was significantly higher, with Australia and the UK experiencing 22% and 19%
respective growth in average rated front-load capacities. Over the period the smallest growth
in rated load of front-loader units was reported in Austria, and even here the increase was
11% over 7 years.

13
However, the apparent significant differences in top and front loader units in Australia should be treated with
caution due to local cultural and test method issues. Refer to Annex 3 for more information.
14
Note that data reported is product weighted load capacities. Insufficient data is available to report changes in
rated load capacities on a sales weighted basis. However, from the limited data that is available, and from
additional anecdotal reports, it is likely that sales weighted growth in product capacities is actually higher than
product weighted growth in rated load capacities, i.e. among the products presented for sale, consumers are by
preference choosing products with larger rated capacities.
`
P a g e | 17 P a g e | 17



Of particular note is the surprisingly large capacity of domestic units in Korea. The rated load
of front-loader machines averages over 10 kg in 2006, with continued rises since this date.

Again this picture of variations in capacity is somewhat complex. However, there are clearly
two messages that policy makers can draw:
The rated load sizes of units are increasing and there is no indication that these
increases in capacity are reaching a plateau in any country. At some point the
physical dimensions of units will create a maximum possible rated load
15
, but it
appears this point is yet to be reached in any market.
The load capacities of top-loading washing machines were generally traditionally
greater than front-loading units. However, this trend has reversed with average front-
loader capacities now larger than average top-loader capacities for all markets where
data is available.
Having established the average load sizes in each country/region, it is now possible to
establish how efficiently units wash each kilogramme of load.

15
The footprint of a unit is the dimensions of the base of the unit (or sometimes defined as the outermost width
and depth dimensions of the product). Note that typical unit footprints vary slightly by region.
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia FL 4.87 5.14 5.07 5.17 5.57 5.77 5.76 6.34 6.41 6.65 6.94 6.98
Australia TL 5.66 5.82 5.96 6.20 6.23 6.27 6.27 6.30 6.35 6.40 6.59 6.78
Austria FL 4.95 5.00 5.04 5.07 5.14 5.31 5.42 5.53 5.66 5.88
Austria TL 4.99 5.00 5.00 5.01 5.00 5.02 5.04 5.10 5.32 5.45
Republic Of Korea FL 10.12 10.20 10.93 10.75
United Kingdom FL 4.97 5.02 5.19 5.27 5.30 5.58 5.62 5.76 5.98
EU FL 4.93 5.05 5.08 5.13 5.24 5.36 5.55 5.75 5.94 6.09
EU TL 4.91 4.90 4.91 4.93 4.98 5.01 5.05 5.15 5.26 5.35
0
2
4
6
8
10
12
A
v
e
r
a
g
e

l
o
a
d

c
a
p
a
c
i
t
y

(
k
g
)
F
r
o
n
t

L
o
a
d
e
r
s

(
F
L
)

a
n
d

T
o
p

L
o
a
d
e
r
s

(
T
L
)
Figure 6. Average product weighted load capacities in kg for both front and top-loader
washing machines.
`
P a g e | 18 P a g e | 18



2.1.5 Product efficiencies
Perhaps the most useful comparison of product energy performance is to examine the
energy used to wash each kilogramme of load, i.e. the amount of energy required to deliver
one unit (kg) of clean laundry. This is the measure of washing machine efficiency as defined
in the original Mapping and Benchmarking washing machine product definition.

Sales weighted normalised average energy efficiencies (kWh/kg) for all washing machines in
each country are shown in Figure 7.
Initially looking at the whole market over the 2000 to2007 period, there has been a
noticeable 18% improvement in efficiency of European washing machines (and in individual
EU countries, 18% in the UK and a significant 33% in Austria). This reflects the combined
action of falls in UEC and increase in volumes. This has resulted in the average unit energy
efficiencies in the EU (and Switzerland) converging to between 80 and 90 Wh/kg. Again the
significant outlier is Australia where, despite an average whole market increase in efficiency
of 40%, the resulting efficiency level of 190 Wh/kg is still more than double the level of any
other country. Obviously this lower level of efficiency in Australia reflects their significantly
higher UEC which is only slightly offset by load capacities marginally larger than those in the
EU countries.

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (40 degC) 0.32 0.30 0.29 0.30 0.32 0.29 0.29 0.24 0.23 0.23 0.21 0.19
Austria (60 degC) 0.12 0.12 0.11 0.11 0.10 0.09 0.09 0.08 0.09 0.08
EU (60 degC) 0.11 0.10 0.10 0.10 0.09 0.10 0.09 0.09 0.09 0.09
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
S
a
l
e
s

w
e
i
g
h
t
e
d

e
f
f
i
c
i
e
n
c
y
(
n
o
r
m
a
l
i
s
e
d

k
W
h
/
k
g

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
Figure 7. Sales weighted normalised average washing machine energy efficiency
(kWh/kg) for all products.
`
P a g e | 19 P a g e | 19



Where information was available, efficiencies were broken into top and front-loader market
segments (Figure 8 and Figure 9 respectively).

In the top-loader segment a similar picture emerges to that in the full market, i.e. Austria and
the EU as a whole have efficiency levels of around 80 Wh/kg, with Australia again being a
significant outlier at 2.5 times these efficiencies. However, in all three cases, efficiency has
improved by 23-25%.

Looking at front-loader units in isolation, improvements in efficiency are similar to those in
top loader units (EU 18%, UK 18% and Austria 33%), but there are significant deviations in
product performance across nations. While the European countries again converge on the
80 and 90 Wh/kg efficiency range, Australia still lags behind other countries at 110 Wh/kg
(albeit this level is more than twice as efficient as Australian top-loaders). Korea performs
extremely well in comparison with the European countries, with unit efficiencies of 60 Wh/kg
(although this is primarily due to the very large unit capacities which are unlikely to be
replicated elsewhere).

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (20+ degC) 0.32 0.31 0.31 0.31 0.34 0.32 0.32 0.29 0.28 0.29 0.28 0.26
Austria (60 degC) 0.12 0.11 0.10 0.10 0.09 0.09 0.09 0.09 0.08 0.08
EU (60 degC) 0.12 0.11 0.10 0.10 0.09 0.11 0.09 0.08 0.08 0.08
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
S
a
l
e
s

w
e
i
g
h
t
e
d

e
f
f
i
c
i
e
n
c
y

o
f

t
o
p

l
o
a
d
e
r
s
(
n
o
r
m
a
l
i
s
e
d

k
W
h
/
k
g

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
Figure 8. Sales weighted normalised average top-loader washing machine energy
efficiency (kWh/kg).
`
P a g e | 20 P a g e | 20



2.1.6 Summary potential improvements in the energy performance of units
Based on the evidence available, Australia has significant potential to adopt policies that
may drive the energy performance of washing machines to align more closely with models
elsewhere. This is particularly the case with top-loading units, but increasingly front-loading
models are diverging from equivalents elsewhere. However, such action may not actually be
necessary given the cultural preference for cold washes in Australia which will result in
actual consumption by the user being significantly below the test values.

Elsewhere there is insufficient evidence to draw reasonable conclusions on the potential for
improvement in product efficiencies given the inherent inaccuracies of the normalisation
approach. However, the evidence does allow the following observations:

1) In almost all markets UEC is beginning to plateau or increase, with recent
improvements in efficiencies primarily related to increasing volumes of the
machines
16
. Therefore, it may be appropriate for policy makers to investigate whether
actual load sizes used by consumers are increasing. If actual load sizes are
increasing in line with the increase in load capacities of units, then clearly the energy
consumption per unit of clean laundry is decreasing as desired. However, if

16
Although this may not be true for the EU. Refer to Footnote 12.
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (40 degC) 0.13 0.17 0.13 0.13 0.11 0.10 0.11 0.10 0.10 0.11 0.11 0.11
Austria (60 degC) 0.12 0.12 0.11 0.11 0.10 0.09 0.08 0.08 0.09 0.08
EU (60 degC) 0.11 0.10 0.10 0.10 0.09 0.10 0.09 0.09 0.09 0.08
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
S
a
l
e
s

w
e
i
g
h
t
e
d

e
f
f
i
c
i
e
n
c
y

o
f

f
r
o
n
t

l
o
a
d
e
r
s
(
n
o
r
m
a
l
i
s
e
d

k
W
h
/
k
g

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
Figure 9. Sales weighted normalised average front-loader washing machine energy
efficiency (kWh/kg).
`
P a g e | 21 P a g e | 21



consumers are selecting larger machines but not increasing actual load sizes, then it
is likely that further improvements in declared product efficiencies may be disguising
stagnation in actual energy consumption per unit of clean laundry. If this is the case,
consideration should be given to limiting unit size and/or capping energy
consumption directly (or a combination of both through amendments to algorithms
that define minimum efficiency levels).

2) In all markets front-loading washing machines now have lower UECs, higher
capacities and better efficiencies than their top-loading competitors. Therefore the
implementation of policy to drive improvement in top-loader energy consumption, or
to encourage consumer switching to front-loader units, would yield significant energy
savings
17
in those countries where top-loader penetration is still high. Such a
conclusion is supported by looking at the distribution of normalised efficiencies of
front and top-loader models in each rated load range in a single year (2007), in a
single market (Australian) (Figure 10). By looking at the single market, unit
efficiencies should be directly comparable for any given rated load capacity. As can
be seen, the efficiencies of front-loading machines (bounded by the red line) are
generally significantly better than top-loader washing machines of the same rated
load capacities (bounded by the white line).

17
Although it is recognised that other performance variables, in particular wash cycle times, may be adversely
affected by such intervention.
0
2
4
6
8
10
12
0 0.1 0.2 0.3 0.4 0.5 0.6
L
o
a
d

c
a
p
a
c
i
t
y

(
k
g
)
Normalised Efficiency (kWh/kg)
Australia 07 Top Loader
Australia 07 Front Loader
Figure 10. Comparative normalised efficiencies of front- (FL) and top- loader (TL)
models in each rated load range for washing machines in the Australian market.
`
P a g e | 22 P a g e | 22



2.2 Observations (including the USA and Canada)
Unit energy consumption information declared in Canada and the USA is based on test
information drawn from a range of cycle conditions (in particular the quantity and
temperature of water used). Due to a shortcoming of the normalisation methodology (which
is based on a single unit energy consumption value for a specific set of operating conditions)
normalised Canadian and USA data results are less comparable than normalised results
from other countries, although the relative positioning of results from Canada and the USA
are highly likely to be comparable. This shortcoming in the normalisation methodology is
compounded by a revision to the Canadian and USA required test method/energy reporting
algorithm in 2004. This revision included changes to the balance of wash temperatures,
such that the hot and warm wash cycles were assigned lower weighting factors after
2004. This revision, on average, will tend to result in higher energy consumption values
being reported prior to 2004 in comparison with those after
18
. Further, the load (tub)
capacities of washing machines in Canada and the USA are declared based on the physical
internal dimensions of the machines which have been converted to kilogramme loads to be
comparable with data reported from elsewhere. Thus, the benchmarked data for Canada
and the USA presented in this section should be interpreted in the context of the
limitations in comparability with washing machine performance from elsewhere as
outlined above.
2.2.1 Unit energy consumption
As with almost all other countries, both Canada and the USA have reported a fall in the sales
weighted manufacturer declared UEC of new washing machines when tested under local
test conditions (Figure 12). This general picture is reinforced when UEC data is normalised
to account for the differences in energy required to heat water for the varying local
test/regulatory requirements, bearing in mind the limitations of the normalisation process for
Canadian and US data (Figure 12. ).

However, in Canada and the USA, this reduction in UEC has been more substantial than
elsewhere, with normalised UECs falling by 71% in Canada, and 67% in the USA, over the
period 2000 to 2007. Nevertheless, this must be viewed in the context that in 2000,
Canadian and US normalised UECs were over twice that of any reporting country, with the
obvious exception of Australia. By 2007 the UECs in Canada and the USA were 0.35 kWh
and 0.4 kWh respectively. These UECs make the Canadian and USA levels marginally the
best of all reporting countries. However, given the limitations of the normalisation process, it
is likely that the Canadian and USA washing machines are actually comparable with the
broad grouping of units elsewhere at around 0.5 kWh.

18
`
P a g e | 23 P a g e | 23



1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (40 degC) 1.57 1.55 1.53 1.62 1.73 1.58 1.58 1.34 1.30 1.33 1.26 1.18
Austria (60 degC) 1.18 1.16 1.11 1.06 1.03 1.00 0.99 0.99 1.00 0.99
Canada CAM (57.2 degC) 2.28 2.24 2.17 2.19 2.14 2.07 1.99 1.81 1.46 1.13 0.99 0.73
Denmark (60 degC) 0.77 0.71 0.78 0.73
USA AHAM (57.2 degC) 2.24 2.21 2.27 2.17 2.20 2.19 2.13 1.97 1.22 1.13 1.18 0.82
EU (60 degC) 1.17 1.10 1.05 1.03 1.00 1.00 0.99 0.95 0.95 0.97
0.0
0.5
1.0
1.5
2.0
2.5
S
a
l
e
s

w
e
i
g
h
t
e
d

c
o
n
s
u
m
p
t
i
o
n

(
d
e
c
l
a
r
e
d

k
W
h

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (40 degC) 1.78 1.75 1.73 1.84 1.97 1.79 1.79 1.48 1.44 1.48 1.39 1.29
Austria (60 degC) 0.57 0.56 0.53 0.51 0.49 0.47 0.47 0.47 0.47 0.47
Canada CAM (57.2 degC) 1.28 1.25 1.21 1.23 1.20 1.15 1.10 1.00 0.79 0.59 0.51 0.35
Denmark (60 degC) 0.35 0.31 0.35 0.32
USA AHAM (57.2 degC) 1.26 1.24 1.27 1.21 1.23 1.23 1.19 1.09 0.64 0.59 0.62 0.40
EU (60 degC) 0.55 0.52 0.50 0.48 0.47 0.50 0.49 0.48 0.49 0.49
0.0
0.5
1.0
1.5
2.0
S
a
l
e
s

w
e
i
g
h
t
e
d

c
o
n
s
u
m
p
t
i
o
n

(
n
o
r
m
a
l
i
s
e
d

k
W
h

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
Figure 12. Sales weighted energy consumption as declared under local test conditions (kWh).
Figure 11. Sales weighted normalised energy consumption (kWh).
`
P a g e | 24 P a g e | 24



The majority of the fall in UEC in Canada and the USA coincides with the announcement
and implementation of revised mandatory Minimum Energy Performance Standards (MEPS)
in both countries in 2004 (and a revision in the labelling requirement in the US), and the
further revisions of these MEPS in 2007
19
(USA) and 2008 (Canada)
20
. However, it should
be noted that the 2004 MEPS revision coincided with a revision to the test method/energy
reporting algorithm which will tend to result in higher UEC values being reported prior to
2004 in comparison with those after. Therefore, the actual degree of product improvement
resulting from the 2004 MEPS may be less than it appears, although still remaining
substantial.

2.2.2 Comparative unit energy consumption based on orientation of the
washing machine
No data was available on the split between UEC for top and front-loading washing machines
in the USA, but this data is available for Canada and is presented alongside similar data for
other reporting countries in Figure 13. and Figure 13

Although the Figures and associated analysis should be considered in the context of the
limitations of comparability, presentation of the Canadian data does provide some interesting
insights:
1) In percentage terms, over the 2000 to 2007 period, the improvement in the UEC of
Canadian top-loader machines (58%) is significantly higher than the improvement in
front-loader machines (42%). However, despite this relative improvement, top-loader
machines still use more than twice the energy of their front-loading equivalents.
2) The significant improvements in Canadian top-loader UECs have only recently
brought them broadly into line with the 0.5 kWh consumptions of equivalent top-
loader machines in Europe.

19
Canada also revised its standard in 2008 to include commercial clothes washers. However, this revision did
not impact the domestic washing machine test method or MEPS level.
20
Refer to Annex 4 for summary details of primary policy actions and associated timescales.
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (40 degC) 0.64 0.86 0.65 0.65 0.59 0.59 0.62 0.62 0.66 0.72 0.75 0.77
Austria (60 degC) 0.57 0.57 0.53 0.51 0.49 0.47 0.47 0.47 0.47 0.47
Canada (57.2 degC) 0.35 0.33 0.35 0.37 0.33 0.31 0.25 0.22 0.19
EU (60 degC) 0.55 0.52 0.50 0.49 0.48 0.49 0.49 0.49 0.50 0.51
0.0
0.5
1.0
1.5
2.0
S
a
l
e
s

w
e
i
g
h
t
e
d

c
o
n
s
u
m
p
t
i
o
n

o
f

f
r
o
n
t

l
o
a
d
e
r
s
(
n
o
r
m
a
l
i
s
e
d

k
W
h

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (20+ degC) 1.81 1.81 1.81 1.93 2.10 1.98 1.98 1.79 1.76 1.84 1.82 1.73
Austria (60 degC) 0.53 0.50 0.46 0.47 0.47 0.46 0.46 0.45 0.44 0.44
Canada (57.2 degC) 1.31 1.32 1.30 1.25 1.18 0.99 0.84 0.76 0.55
EU (60 degC) 0.57 0.53 0.50 0.49 0.47 0.47 0.47 0.44 0.44 0.46
0.0
0.5
1.0
1.5
2.0
S
a
l
e
s

w
e
i
g
h
t
e
d

c
o
n
s
u
m
p
t
i
o
n

o
f

t
o
p

l
o
a
d
e
r
s
(
n
o
r
m
a
l
i
s
e
d

k
W
h

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
Figure 14. Sales weighted normalised energy
consumption for Top Loader Machines (kWh).
Figure 13. Sales weighted normalised energy
consumption for Front Loader Machines (kWh).
`
P a g e | 25 P a g e | 25



3) Despite the improvements in both front and top-loader performance, individually
neither account for the 71% improvement in the overall UEC of new washing
machines in Canada over the 2000 to 2007 period. Much of this overall market
improvement is the result of a market switch from the higher consuming top-loaders,
to the lower consuming front-loaders the market share of front-loaders in 2001 was
16%, but this had risen to 57% by 2008.
2.2.3 Trends in rated load sizes and product efficiencies
Contrary to belief among many consumers, as Figure 15 illustrates, Canadian washing
machines do not have markedly larger capacity than European models (despite a slightly
larger footprint) and are significantly smaller than comparable Australian units. In fact, given
the uncertainty in the approach to normalisation of load size, there is a possibility that the
sizes shown for Canadian washing machines is somewhat exaggerated and their load
capacities are smaller than European models
21
. However, as elsewhere, the trend is for
increasing load capacities, particularly in front-loading units
22
. Over the 2003 to 2009 period,

21
Although it should be noted that Canadian capacities are based on the physical dimensions of the unit whereas
European capacities are based on manufacturer declared values which are used to define the load under test
so comparison is not direct.
22
Note that data reported is product weighted load capacities. Insufficient data is available to report changes in
rated load capacities on a sales weighted basis. However, from the limited data that is available, and from
additional anecdotal reports, it is likely that sales weighted growth in product capacities is actually higher than
product weighted growth in rated load capacities, i.e. among the products presented for sale, consumers are by
preference choosing products with larger rated capacities.
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia FL 4.87 5.14 5.07 5.17 5.57 5.77 5.76 6.34 6.41 6.65 6.94 6.98
Australia TL 5.66 5.82 5.96 6.20 6.23 6.27 6.27 6.30 6.35 6.40 6.59 6.78
Canada FL 4.21 4.67 5.06 5.18 5.26 5.46 5.95
Canada TL 5.57 5.64 5.76 5.64 5.66 5.66 5.76
Republic Of Korea FL 10.12 10.20 10.93 10.75
EU FL 4.93 5.05 5.08 5.13 5.24 5.36 5.55 5.75 5.94 6.09
EU TL 4.91 4.90 4.91 4.93 4.98 5.01 5.05 5.15 5.26 5.35
0
2
4
6
8
10
12
A
v
e
r
a
g
e

l
o
a
d

c
a
p
a
c
i
t
y

(
k
g
)
F
r
o
n
t

L
o
a
d
e
r
s

(
F
L
)

a
n
d

T
o
p

L
o
a
d
e
r
s

(
T
L
)
Figure 15. Average product weighted load capacities in kg for both front and top-loader
washing machines.
`
P a g e | 26 P a g e | 26



the average load capacity of front-loading units has increased from 4.2 kg to 6 kg. Front-
loading units now have average capacities significantly larger than the higher energy
consuming top-loading machines.

As noted previously, perhaps the most useful comparison of product energy performance is
to compare the energy used to wash each kilogramme of load, i.e. the amount of energy
required to deliver one unit (kg) of clean laundry. Noting again that caution should be applied
when comparing the Canadian and USA results with those from elsewhere, Figures
illustrating comparisons of whole market product efficiency (Figure 16), and comparisons for
Canadian top-loader (Figure 17. ) and front-loader (Figure 17) units are included for
completeness.

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (40 degC) 0.32 0.30 0.29 0.30 0.32 0.29 0.29 0.24 0.23 0.23 0.21 0.19
Austria (60 degC) 0.12 0.12 0.11 0.11 0.10 0.09 0.09 0.08 0.09 0.08
Canada CAM (57.2 degC)* 0.24 0.23 0.23 0.24 0.22 0.21 0.21 0.21 0.19 0.10 0.10 0.07 0.07 0.07
USA AHAM (57.2 degC) 0.24 0.23 0.24 0.22 0.23 0.22 0.21 0.19 0.11 0.10 0.11 0.07
EU (60 degC) 0.11 0.10 0.10 0.10 0.09 0.10 0.09 0.09 0.09 0.09
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
S
a
l
e
s

w
e
i
g
h
t
e
d

e
f
f
i
c
i
e
n
c
y
(
n
o
r
m
a
l
i
s
e
d

k
W
h
/
k
g

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
* = product weighted
Figure 16. Sales weighted normalised average washing machine energy efficiency
(kWh/kg) for all products.
`
P a g e | 27 P a g e | 27



2.3 Key issues for policy makers
Based on the evidence available, policy makers may wish to note the following:
The combination of MEPS and labelling has had a very strong impact in Canada and
the USA (although the biggest market response seems to align primarily with the
introduction/revision of MEPS and the associated labelling revisions). Ultimately
Canadian and US unit energy consumptions are now (likely) to be broadly in line with
most European countries despite previously being at significantly higher initial levels.
Therefore, it is reasonable to conclude that challenging and regularly revised MEPS
are a highly effective method of reducing consumption.
The impact of labelling within the EU countries, and the associated voluntary
agreement with industry, has had a mixed effect. Across the EU as a whole and
within Austria, there has been a significant market response. However, within the UK
improvement has been minimal
23,24
. Therefore policy makers should be aware of the
potential improvements that can be achieved through the combined application of
labelling and voluntary agreements. However, ongoing monitoring of the market
should be undertaken to ensure policy impacts are occurring in all market segments
and, where this is not the case, revisions to the agreement/policy should be
considered.

23
Unfortunately there is insufficient information available at this time to account for this difference in consumer
purchasing patterns (eg whether the difference is caused by pricing strategies, consumer information/preference,
model availability, etc).
24
Note that the improvements across of the EU have reduced in the most recently reported years. There is some
anecdotal evidence that this is due to a lack of a revision of the labelling categories. As a large proportion of units
have reached the highest A level, there is no visible sign to the consumer that a product has improved past this
level. Thus, manufacturers have no incentive to improve products, or if they do, no incentive to report this
improvement until the labelling categorisation system is revised (there is some limited test data to support the
latter thesis).
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (40 degC) 0.13 0.17 0.13 0.13 0.11 0.10 0.11 0.10 0.10 0.11 0.11 0.11
Austria (60 degC) 0.12 0.12 0.11 0.11 0.10 0.09 0.08 0.08 0.09 0.08
Canada (57.2 degC)* 0.08 0.07 0.05 0.05 0.04 0.04 0.03
EU (60 degC) 0.11 0.10 0.10 0.10 0.09 0.10 0.09 0.09 0.09 0.08
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
S
a
l
e
s

w
e
i
g
h
t
e
d

e
f
f
i
c
i
e
n
c
y

o
f

f
r
o
n
t

l
o
a
d
e
r
s
(
n
o
r
m
a
l
i
s
e
d

k
W
h
/
k
g

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
Figure 18. Sales weighted normalised average top-
loader washing machine energy efficiency (kWh/kg).
Figure 17. Sales weighted normalised average front-
loader washing machine energy efficiency. (kWh/kg)
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (20+ degC) 0.32 0.31 0.31 0.31 0.34 0.32 0.32 0.29 0.28 0.29 0.28 0.26
Austria (60 degC) 0.12 0.11 0.10 0.10 0.09 0.09 0.09 0.09 0.08 0.08
Canada (57.2 degC)* 0.23 0.21 0.12 0.12 0.09 0.09 0.10
EU (60 degC) 0.12 0.11 0.10 0.10 0.09 0.11 0.09 0.08 0.08 0.08
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0.50
S
a
l
e
s

w
e
i
g
h
t
e
d

e
f
f
i
c
i
e
n
c
y

o
f

t
o
p

l
o
a
d
e
r
s
(
n
o
r
m
a
l
i
s
e
d

k
W
h
/
k
g

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
`
P a g e | 28 P a g e | 28



In the one country reporting which relies solely on labelling as its policy intervention
(Australia), improvements have been seen in overall average unit energy
consumption of washing machines, but the performance of these units still lags
significantly behind all other participating countries (Australian unit energy
consumption is more than twice that of the country with the next highest unit
consumption, and the consumption of Australian front-loading units is actually rising,
which is at odds with all other countries). However, the significant use of cold water in
washes by Australian consumers (which is not reflected in the test methodology) may
negate the need for policy intervention to address this apparent difference.
The implementation of policy to drive improvement in the energy performance of top-
loading washing machines, or to encourage consumer switching from top-loader
machines to their front-loader competitors, would yield significant energy savings in
those countries where top-loader penetration is still high. However, policy makers
should be aware that such action may adversely affect other performance variables,
in particular wash cycle times (see the following section).
The rated load sizes of washing machines are increasing in almost all jurisdictions
and there is no indication that these increases are reaching a plateau in any country.
This ongoing increase in rated capacity is at least partly responsible for increasing
product efficiency (more so in recent years where the improvement in unit energy
consumption is beginning to stagnate in some regions). Hence policy makers may
wish to investigate whether the actual size of loads washed by consumers is
increasing in line with increasing machine sizes. If actual load sizes are not
increasing significantly, it is likely that further improvements in declared product
efficiencies due to increasing product sizes may be disguising stagnation in actual
energy consumption per unit of clean laundry. If this is the case, consideration should
be given to limiting unit size and/or capping energy consumption directly (or a
combination of both through amendments to algorithms that define minimum
efficiency levels).
The load capacities of top-loading washing machines were traditionally greater than
front-loading units. However, this trend has reversed, with average front-loader
capacities now larger than average top-loader capacities for all markets where data
is available.
`
P a g e | 29 P a g e | 29



3 Non energy performance variables
3.1 Observations
3.1.1 Water consumption
The water consumption of individual washing machines is labelled in almost all countries in
conjunction with the energy consumption/efficiency label; or as a stand-alone water
efficiency label in Australia (the exception being Canada).

Water consumption has been falling in almost all countries over an extended period,
although recently such consumption has begun to plateau (albeit at significantly differing
levels) with slight recent increases being experienced in some countries (see Figure 19).
However, water efficiency (defined as water consumption in litres per kg of laundry) has
continued to reduce in all countries for all years as would be expected given the increasing
load sizes (Figure 20). It is to be expected that water consumption would reduce in a
proportion directly linked to energy consumption (when non cold washes are used).
Reported average water efficiency reductions do indeed almost exactly match reductions in
average unit energy efficiency in each market. However, there will come a point where
further reductions in water consumption can no longer be sustained without significant
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia FL 98 91 81 69 67 71 73 74 71 71 67 67
Australia TL 151 141 143 149 138 144 144 139 124 119 110 108
Austria FL 56 54 53 51 50 49 49 49 50 50
Austria TL 57 56 53 50 48 50 52 50 50 49
Canada FL 65 60 55 53 51
Canada TL 144 125 130 128 128
Republic Of Korea FL 96 94 110 105
United Kingdom FL 69 63 58 57 54 52 54 54
USA Energy Star FL 63 62 61
China FL 68 71 65
China TL 159 167 164
EU FL 53 49 48 46 46 47 49 49 50 50
EU TL 60 58 56 53 52 50 49 48 48 48
0
20
40
60
80
100
120
140
160
180
T
o
t
a
l

w
a
t
e
r

c
o
n
s
u
m
p
t
i
o
n

b
y

f
r
o
n
t

l
o
a
d
e
r
s
(
F
L
)

a
n
d
T
o
p

L
o
a
d
e
r
s

(
T
L
)

i
n

a
s
t
a
n
d
a
r
d

w
a
s
h

(
l
i
t
r
e
s
)
Figure 19. Total water consumption of front and top-loading washing machines in a
standard wash using local test methodologies (litres).
`
P a g e | 30 P a g e | 30



deterioration in wash performance and/or rinse effectiveness; or without increased cycle
time. It is possible that such a point is approaching as water consumption and water
efficiency are both beginning to plateau. Therefore, in countries where water and/or energy
consumption/efficiency will continue to be a focus of policy intervention, policy makers
should consider increasing vigilance of wash and rinse performance to ensure their policy
intervention will not result in impaired unit performance to which consumer reaction may be
negative.

3.1.2 Wash quality
Given the observation on water consumption and potential impact on wash quality, evidence
to date suggests that so far the reductions in water consumption have had no adverse
impact on wash quality, at least in those countries where wash quality is measured. While
absolute wash quality is not comparable, Figure 21 illustrates the relative changes in wash
performance in each market relative to the first year of data availability in that market all of
which are positive.
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia FL (40 degC) 20.1 17.8 15.9 13.3 12.0 12.3 12.6 11.7 11.2 10.7 9.6 9.6
Australia TL (40 degC) 26.6 24.2 24.0 24.1 22.1 22.9 22.9 22.1 19.5 18.5 16.8 16.0
Austria FL (60 degC) 11.4 10.9 10.5 10.0 9.7 9.3 9.0 8.8 8.8 8.4
Austria TL (60 degC) 11.3 11.2 10.6 10.0 9.6 10.0 10.2 9.7 9.4 9.1
Canada FL (57.2 degC) 12.9 11.7 10.5 9.7 8.5
Canada TL (57.2 degC) 25.0 22.1 22.9 22.7 22.2
Republic of Korea FL (37 degC) 9.5 9.2 10.1 9.8
UK FL (60 degC) 13.8 12.2 11.1 10.7 9.8 9.3 9.4 9.0
USA Energy Star FL (57.2 degC) 11.3 11.2 10.6
China FL (unknown) 11.6 11.7 10.3
China TL (unknown) 24.9 24.2 24.4
EU FL (60 degC) 10.7 9.7 9.4 9.0 8.7 8.8 8.8 8.5 8.4 8.3
EU TL (60 degC) 12.2 11.9 11.4 10.7 10.4 10.1 9.8 9.4 9.2 9.0
0
5
10
15
20
25
30
W
a
t
e
r

e
f
f
i
c
i
e
n
c
y

(
l
i
t
r
e
s
/
k
g
)
ALL DATA IS ILLUSTRATIVE
Figure 20. Water efficiency of front and top-loading washing machines in a standard
wash using local test methodologies (litres).
`
P a g e | 31 P a g e | 31



3.1.3 Spin effectiveness
Spin effectiveness is a measure of how much residual moisture is left in clothes at the end of
the washing cycle. Such a measurement is of particular importance in those countries where
the use of laundry dryers is significant, as any additional moisture left in the laundry will
result in increases in drying energy.

Again local test measurements are non-comparable, but Figure 22 illustrates the changes in
spin effectiveness in each market relative to the first year of data availability in that market.
In line with the improvements in wash quality, all reporting countries have shown an
improvement in spin effectiveness (reduction in residual moisture content). However, as
other options to reduce the energy consumption of units appear to be reducing (as
evidenced by unit energy consumption appearing to reach a plateau), there may be a
tendency for manufacturers to sacrifice spin performance in order to reduce overall unit
consumption. Therefore, policy makers may wish to keep a watching brief on spin efficiency
to ensure consumers remain satisfied with spin performance, and to ensure that
improvements in washing machine energy consumption through reduced spinning are not
resulting in significantly greater increases in consumption in the post-wash drying of the
laundry.

Figure 21. Wash quality relative to performance in the first year data is available.
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia 1.00 1.01 1.01
Austria 1.00 1.00 1.01 1.01 1.01 1.01 1.02 1.02 1.02 1.02
United Kingdom 1.00 1.01 1.03 1.04 1.04 1.04 1.04 1.04 1.04
EU 1.00 1.00 1.01 1.01 1.02 1.02 1.02 1.03 1.03 1.03
0.99
1.00
1.01
1.02
1.03
1.04
1.05
W
a
s
h

q
u
a
l
i
t
y

i
n
d
e
x

r
e
l
a
t
i
v
e

t
o

e
a
r
l
i
e
s
t

y
e
a
r

d
a
t
a

i
s

a
v
a
i
l
a
b
l
e

(
e
a
r
l
i
e
s
t

y
e
a
r

=
1
.
0
)
`
P a g e | 32 P a g e | 32



Figure 22. Spin effectiveness relative to performance in the first year data is available.
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia 1.00 0.97 0.95 0.94 0.90 0.91 0.91 0.88 0.86 0.87 0.82 0.82
Austria 1.00 0.98 0.96 0.95 0.94 0.93 0.93 0.92 0.91 0.90
United Kingdom 1.00 0.98 0.97 0.97 0.97
EU 1.00 1.00 0.99 0.98 0.95 0.94 0.92 0.91 0.91 0.89
0.80
0.85
0.90
0.95
1.00
1.05
R
e
d
u
c
t
i
o
n

i
n

r
e
s
i
d
u
a
l

w
a
t
e
r

c
o
n
t
e
n
t
I
n
d
e
x

r
e
l
a
t
i
v
e

t
o

e
a
r
l
i
e
s
t

y
e
a
r

d
a
t
a

i
s

a
v
a
i
l
a
b
l
e
(
e
a
r
l
i
e
s
t

y
e
a
r

=
1
.
0
)
`
P a g e | 33 P a g e | 33



Annex 1 Terminology used
The following lists some of the terminology used within this benchmarking document. It does
not attempt to provide a full listing of all terminology, but rather to provide a summary of
terminology most frequently used and/or terminology used in a context with a meaning that
is less well known or different to its more common usage.

Front-loading washing
machines
Washing machines where the load is inserted through a
door in the front of the machine (often referred to as
horizontal axis washing machines)
Load The size and type of laundry being washed by the
machine as defined in local test conditions
Product weighted data Data that has been weighted in line with the number of
individual products reportedly available to the consumer
Rated load The maximum size of load that can be washed by the
machine as declared by the manufacturer (note that the
benchmarking uses kilogrammes to define load although
Canada and the USA define the capacity of the drum
refer to Annex 3 for details of the conversion used)
Rinse effectiveness A measure of how effectively detergent has been
extracted from the load at the end of the washing cycle
(values quoted use local test declaration unit)
Sales weighted Data that has been weighted in line with the number of
total sales of individual products
Spin effectiveness A measure of how much water has been extracted from
the load at the end of the washing cycle (values quoted
use local test declaration unit)
Top-loading washing
machines
Washing machines where the load is inserted through a
door on the top of the machine (often referred to as
vertical axis washing machines)
Unit Energy
Consumption (UEC)
The energy consumption of the washing machine under
local test conditions (or following normalisation) in kWh
Unit energy efficiency The energy consumption per kilogramme of laundry
washed under local test conditions (or following
normalisation) in kWh/kg
Washing cycle time The average period of time to complete the test cycle as
declared by the manufacturer
Wash quality A measure of how effectively the washing machine has
cleansed the dirt from the load at the end of the washing
cycle (values quoted use local test declaration unit)
`
P a g e | 34 P a g e | 34



Washing temperature The nominal temperature of the water as defined in local
test conditions
Water consumption The total water consumption (both hot and cold) of the
washing machine under local test conditions (or following
normalisation) in litres
Water efficiency The water consumption per kilogramme of laundry
washed under local test conditions in litres/kg

`
P a g e | 35 P a g e | 35



Annex 2 Framework for grading mapping and
benchmarking outputs
In order for the Mapping and Benchmarking Annex to provide transparency regarding the
degree of reliability that can be attributed to the results produced by the Annex, a
framework has been developed that allows the grading of benchmarking outputs. This
grading is based on a three part scale of robust, indicative and illustrative. This grading is
applied to both the initial data input and any manipulations that are required to present the
data in a consistent form in the country mappings, and to the subsequent manipulations of
that data in order to make it comparable with datasets from other countries/regions during
the benchmarking process. While expert opinion is used to formulate the specific grading
allocated to individual datasets or outputs, this expert opinion is formed with the following
framework.
Grading of data/mapping outputs
Robust where typically:
The data are largely representative of the full market and
The data include at least a significant element of individual product data and
The data are from known and reliable sources and
Test methodologies are known and reliable and
Any data manipulations are based on solid evidence and should not unduly distort
results.
Conclusions from such datasets are as reliable as reasonably possible within boundaries of
the Annex operation.
Indicative where typically:
Datasets may not be fully representative of the markets (but do account for a
majority, ideally a known and understood majority) and/or
Any data manipulation used includes some assumptions or unavoidable
approximations that could unintentionally reduce accuracy.
Accuracy is, however, judged such that meaningful but qualified conclusions could be drawn.
Illustrative where typically:
One or more significant parts of a dataset is known to represent less than a majority
of the full market or
Test methodologies used to derive data are not known or
Test methodologies used to derive data are known but could lead to significant
differences in outcome or
Data manipulations for the analysis contain an element of speculation or significant
assumption or
Conflicting and equally valid evidence is available.
`
P a g e | 36 P a g e | 36



Rather than being rejected completely, perhaps because the flaws in the data are at least
consistent, such data could provide some insight into the market situation and so are worth
reporting, but results must be treated with caution.
Grading of comparison between country outputs (benchmarking)
Robust where typically:
The data sources being compared are each largely robust and
No data manipulations for benchmarking were necessary; or if manipulations were
used they were based upon solid evidence and should not distort results.
Conclusions from comparisons within and between such datasets are as reliable as
reasonably possible within boundaries outlined above.

Indicative where typically:
Datasets being compared are themselves only indicative and/or
Any data manipulation used for benchmarking includes some assumptions or
unavoidable approximations that could unintentionally reduce accuracy and/or
For any other reason(s) subsets of the data may not be strictly comparable which
leads to some distortion.
However, accuracy is such that meaningful but qualified conclusions could be drawn.
Illustrative where typically:
One or more significant parts of the datasets are themselves illustrative and/or
Data manipulations for the benchmarking process contain an element of speculation
or significant assumption.
Rather than being rejected completely, perhaps because the flaws in the data are at least
consistent, such data could provide insight into the market situation and so are worth
reporting, but results must be treated with caution.

`
P a g e | 37 P a g e | 37



Annex 3 Categorisation of original datasets, approach to
normalisation and associated cautions
Original data quality
Significant efforts have been made by all participating parties to obtain and process data
from a range of sources and to ensure the integrity of the data supplied. However, inevitably
there have been some difficulties sourcing information for all countries/regions, and indeed
in sourcing all information from individual countries/regions even where this information
exists. Therefore, the specific nature of each dataset is different. For example, some
datasets are based on detailed information on individual models across a whole market;
others are based on averages of aggregated data from the whole market, and some data
sets are selections/samples which may be representative of the market as a whole or just a
subset of the market (for example the best performing products).

Further, in some cases, data manipulation is required to make the material comparable
between countries (for example, the conversion of North American unit capacities defined by
the overall drum size in gallons or litres to the equivalent load capacity in kilogrammes used
elsewhere
25
).

Thus, to provide readers with an indication of the relative reliability of a particular dataset
within the context of the other data being presented, the Mapping and Benchmarking Annex
has developed the Framework for Grading Mapping and Benchmarking Outputs
26
. This
framework enables the allocation of gradings based on a robust, indicative and illustrative
scale. The original data received (including any manipulations necessary
27
) have been
classified based on this framework, with the associated gradings shown in Figure 24.
Benchmarking information
To enable the comparison or benchmarking of product performance between countries, the
Mapping and Benchmarking Annex compares products based on their performance when
undergoing the standard test defined in the local test methodology. Differences in the local
test methodologies are then normalised in an effort to make original data supplied from
each region comparable with data from elsewhere.

However, such an approach leads to some inherent issues of which readers should be
aware:

25
Data on the capacity of Canadian washing machines was supplied as a drum volume. To convert this to a load
in kg, the maximum capacities in table 3 (CSA/C360-03 standard) were used to create an equation from which an
equivalent load in kg could be calculated for any given declared capacity. The resulting formula used was Load
(kg) = 0.0659 Container volume (L) 0.0137. No capacity data was available on machines from the USA.
26
This Framework is generally used across all Mapping and Benchmarking outputs (refer to Annex 2).
27
All manipulations of individual datasets are detailed in the associated country/region mapping at
`
P a g e | 38 P a g e | 38



Normalisation/conversion factors are not 100% accurate. This is a particular issue for
washing machines where a number of variables (water consumption and
temperature, wash quality, drying performance, etc) are interdependent and affect
overall energy consumption and unit efficiencies.
Test methodologies (in particular wash temperatures) may not reflect actual
consumer usage patterns and/or units may not comply fully with the stated
requirements of the test.
These issues are investigated below and specific areas of importance highlighted to the
reader.
The context of normalisation of data
Interrelationship of performance variables
The energy performance of washing machines is interdependent on a range of other
variables including:
Washing temperature, typically the higher the washing temperature, the cleaner the
load (although this variation is more complex now given the advent of enzyme
washing powders);
Washing cycle time, typically the longer the wash time, the cleaner the wash for a
given quantity of water/energy;
Wash quality, typically the higher the wash quality, the more energy (through some
combination of increased water temperature, increased agitation of the load and an
extended wash time) is required to deliver the improved performance;
Spin effectiveness, typically the drier the clothes following the spin cycle, the more
energy has been expended to extract the water;
Rinse effectiveness, typically units that deliver better rinse performance (the
removal of detergent and softener residue) consume more water or have longer
wash cycles;
Type of load, dense cotton material has a different wash and spin requirement to the
more gentle action required for woollens;
Size of load, with a given washing container size, typically the greater the size of the
washing load the lower the washing performance as the degree of agitation is
reduced;
Other variables including the water hardness, specific detergent used, etc all impact
on the performance of one or more of the variables outlined above
Even where the same test method is used, there is a great deal of uncertainty over how
each of these factors interrelate for an individual machine. This means the development of
normalisation factors related to an individual variable (for example energy consumption) will
have inherent shortcomings as it does not factor in the associated positive or negative
impact of other performance variables. For example, a unit that uses slightly more energy
than a competing unit may have overall better performance in terms of wash quality, dryness
of the wash at cycle completion, use of water and other resources such as detergent, etc.
`
P a g e | 39 P a g e | 39



Hence the unit that uses slightly more energy may be considered to be the more resource
efficient for a given level of performance.

Within the context of the specific datasets received, these issues are further compounded by
the provision by some countries of product level data which allow analysis of the interrelation
of many variables for individual machines, and the supply of aggregated market data from
elsewhere which makes such analysis impossible.
Differences in local test methodologies and regulatory requirements
Having established that even where testing requirements are identical, variables are
interrelated, it is evident the development of conversion factors based on a single variable is
challenging. This is further complicated by the difference in:
Testing methodologies between countries/regions and the associated mandatory
conditions of the test (type of load, water temperature, detergent, etc);
Regulations between countries/regions that define minimum performance standards
for one or more of the variables; the requirement to report/declare the performance
related to one or more of the variables; or no requirement to measure or report
anything other than unit energy performance.
To provide a simplified example, local requirements in country A may specify a test wash
temperature of 40
o
C with no requirements to report wash performance or spin effectiveness.
The requirements in country B may be to test at 60
o
C, and the units have a requirement to
achieve (or report on the label) a level of wash performance and spin effectiveness. Clearly
the difference in wash temperature will affect the amount of energy consumed by a machine.
However, even if there is a correction/normalisation made for the differences in energy
required to heat the water, overall consumption of energy for a machine in country B may be
higher than that in country A in order to meet the minimum specified performance level or to
present premium performance characteristics to the consumer via the label information
(something that is not required in country A and therefore suppliers have no incentive to
provide).

A summary of the differing wash test requirements, mandatory performance requirements
and labelling declarations is given below
28
:
Australia defines a minimum wash performance and rinse effectiveness (for warm
wash) with energy and water consumption as labelled variables. The warm wash
temperature test has water inlet temperature of 20
C and a 40
C wash. Australian
normalisation has been undertaken based on the nominal warm wash temperature
defined in the local standard. However consumers are known to perform a large
number of washes in cold water (at the temperature of water intake or similar).
Hence, comparative benchmarking of Australian data, particularly that data

28
For more details on local test requirements, mandatory performance requirements and labelling declarations,
please refer to individual country/region mappings provided at http://mappingandbenchmarking.iea-
4e.org/matrix?type=product&id=3
`
P a g e | 40 P a g e | 40



associated with top-loading washing machines, should be treated with extreme
caution
29
.
.

China requires the declaration and labelling of energy consumption only. Nominal
test temperatures are water inlet temperature of 20
C and a 40
C wash. However, the

labelling requirements are based on the specific functionality of the machine which is
not recorded in the aggregated market data provided. Therefore, data provided
indicates that average washing machines consume less energy than that required
simply to heat water through the temperature change required for testing. As it has
been impossible to obtain a dataset that accounts for the differences in functionality,
Chinese data has been excluded from the Benchmarking Analysis.
EU countries have had, until recently, voluntary minimum energy performance
requirements and mandatory labelling of energy, wash quality, spin effectiveness and
noise (current regulatory revisions involve the removal of wash quality as a labelled
item), but other variables are not regulated or reported to the consumer. The wash
temperature test is 60
C with a cold water inlet of 15
C (recent revisions change wash

temperatures to a weighted average of wash temperatures but these revisions occur
after the last date of data analysed).
Switzerland has identical labelling requirements to the EU from 2003 onward, with
the additional requirement for mandatory energy performance levels from 2010
onward (equivalent to A on the EU energy label).
The Republic of Korea has mandatory energy performance requirements and
labelling of wash quality, rinse effectiveness and spin effectiveness. Different testing
methodologies are used for front-loader units (similar to the EU but with water inlet
temperature of 15
C and a 40
C wash) and top-loader units (water inlet temperature

and wash of 15
C).
US/Canada have mandatory energy and water consumption requirements (water
currently only in the USA) but other variables are not regulated and have no labelling
requirement.
Wash temperature and water inlet temperatures are nominally 57.2
C and 15.6
C
respectively. However, declared energy and water consumption is based on the
average machine consumption over a range of cycle conditions (in particular the
quantity and temperature of water used). A shortcoming of the normalisation
methodology used is that it is based on a unit energy consumption value for a
specific set of operating conditions. As the normalisation of data from Canada and
the USA is based on this average unit consumption at the nominal operating
temperatures rather than a value at a specific operating condition, the resulting
normalised data will be less comparable than normalised results from other

29
Refer to Annex 3, in particular the section Test Declarations, Actual Water Temperature and Normal Usage
for details of the sensitivity of the normalisation approach and benchmarking outcomes to temperature variation.
`
P a g e | 41 P a g e | 41



countries, although the relative positioning of results from Canada and the USA are
highly likely to be comparable.
Further, in 2004 the Canadian and USA test method/energy reporting algorithm was
revised. This revision included changes to the balance of wash temperatures, such
that the hot and warm wash cycles were assigned lower weighting factors after
2004. This revision, on average, will tend to result in higher energy consumption
values being reported prior to 2004 in comparison with those after
30
.
Additionally, the load (tub) capacities of washing machines in Canada
31
are declared
based on the physical internal dimensions of the machines. These have been
converted to a nominal kilogramme load equivalent to those elsewhere based on a
standardised conversion of tub capacity to kilogrammes provided in the national
standards. Such load conversion may not be accurate for all machines.

Thus, benchmarked data from Canada and the USA is presented in separate
sections. The results presented in these sections should be interpreted in the
context of the limitations in comparability with washing machine performance
from elsewhere as outlined above.
Approach to normalisation
While recognising that key performance characteristics (energy consumption, load type and
size, wash quality, etc.) are all intrinsically linked:
No public domain information has been identified that has allowed the Mapping and
Benchmarking Annex to compare washing machines of differing performance across
all the performance variables;
There is limited public domain information on the impact of the various testing
methodologies on the reported overall performance of individual units, or indeed, to
convert the individual performance characteristics where they are measured
differently (for example the spin effectiveness using the European test method and
load compared with test methods and loads elsewhere);
Original data available to the Annex has varying levels of detail ranging from product
level information on almost all performance variables, to aggregated market
averages of energy performance only.
Therefore, the benchmarking analysis and reporting is restricted to:
Normalisation of energy consumption based on a correction for nominal
test/performance standard water inlet temperatures and wash temperatures (see
below);

30
Appendix J1 introduced the drying energy into the MEF metric, the post-2004 energy consumption values used
for this report do not incorporate drying energy (i.e., they represent only machine electrical + hot water energy).
31
A similar conversion would have been required for washing machines in the USA. However, load capacity data
from the USA was not available.
`
P a g e | 42 P a g e | 42



Data that is as declared under local test methodologies for all the individual
performance characteristics of the washing machines other than energy, i.e. no
normalisation was undertaken to account for variations in the measurement of these
variables between testing regimes, nor for the associated impact on energy
consumption.
Methodology for normalisation of energy consumption based on local test
methodologies
Energy consumed by washing machines can be broken down broadly into two elements:
Mechanical energy primarily used to pump water in and out of the washing machine
and to provide the agitation of the wash through drum rotation, impeller action or
other means, plus the energy used in any spinning to remove water from the laundry
at various points in the cycle;
Water heating energy which is the energy required to raise the water from inlet
temperature to the specified wash temperature.
None of the data reported to the Annex separate the energy used in these two elements, nor
is such reporting required in any of the local regulations reviewed. However, during the
development of the process for normalisation of unit energy consumption, the Australian
Government provided access to significant quantities of testing information that indicated the
mechanical energy used by a washing machine was typically in the range 140-180 W cycle,
i.e. typically between 10% and 20% of total washing machine consumption. Consequently,
even a significant reduction (for example 20%) in this mechanical energy consumption would
lead to relatively minor changes in the overall energy consumption of the washing machine.
Therefore, the Annex participants agreed to focus attention on normalising for variations in
nominal test temperatures between countries (hence focusing attention on water heating
energy) by assuming a nominally fixed mechanical energy consumption of 150 Wh/cycle for
all washing machines.

Using this approach, even where the total amount of water heated in the cycle is unknown
(e.g. many units use an undeclared combination of hot and cold water during the washing
and rinsing cycles), the total energy consumed in water heating can be deduced by:

Water heating energy = Total declared energy consumption 150 Wh

Normalisation can then be achieved through:

Normalised energy consumption = Water heating energy consumption x (test
methodology wash-inlet temperature)/(nominal wash-inlet temperature)

where nominal temperatures are:
Inlet temperature = 15
C
Wash temperature = 40
C
(see Differences in local test methodologies and regulations above for local test
temperature).
`
P a g e | 43 P a g e | 43



The following example illustrates the normalisation process:

Declared energy consumption of washing machine = 1 kWh/cycle

By assuming mechanical energy consumption of 150 Wh/cycle

Water heating energy = 1000 Wh 150 Wh = 850 Wh

Now assuming:
Test method water inlet temperature = 20
C
Test method water wash temperature = 60
C

Then
Normalised water heating energy = 850 x (40-15)/(60-20) = 850 x 0.625
Normalised water heating energy = 850 Wh x 0.625 = 531.25 Wh

And
Total normalised energy consumption of washing machine = 531.25 Wh + 150 Wh
Total normalised energy consumption of washing machine = 681.25 Wh

Clearly this approach has flaws of which readers should be aware when reviewing inter-
country comparisons. In particular:
Washing machines are optimised to local conditions. Therefore, simple correction of
water temperature is not thoroughly robust as units tested at a lower temperature
(typically) require less water to achieve the same wash performance when operated
at higher wash temperatures. Hence normalisation using the original water quantity is
somewhat misleading.
The normalisation of Canadian and US data is based on the average machine
consumption over a range of cycle conditions (in particular the quantity of water
used). Thus, benchmarked data from Canada and the USA is presented in
separate sections and should be interpreted in the context of the limitations in
comparability with washing machine performance from elsewhere as outlined
above.
Test declarations, actual water temperature and normal usage
The difference in the requirements of individual testing methodologies and the normal use
of washing machines by consumers adds one further complication to the interpretation of
benchmarked data. To provide an illustration, Figure 23 compares the normalised unit
energy consumption results for Australia using Australian nominal test temperatures (40
C);
the equivalent normalised results for the EU as a whole using EU nominal test temperatures
(60
C); and the same EU data using a nominal wash temperature of 43
C (which is both
closer to the average wash temperature used by consumers, and also closer to the actual
`
P a g e | 44 P a g e | 44



average hot water temperature achieved by a small number of units undergoing the
compliance testing in the UK).

As can be seen, the comparative difference in performance between EU and Australian
washing machines falls sharply when the 43
C nominal temperature is used for EU

machines. This demonstrates the sensitivity of this normalisation approach to nominal test
temperatures. This is known to be of particular relevance to benchmarking data for Australia
where normalisation has been performed to the nominal warm wash test temperature,
however consumers are known to perform a large number of washes in cold water (at the
temperature of water intake or similar). Hence, comparative benchmarking of Australian
data, particularly that data associated with top-loading washing machines, should be
treated with caution.

Important cautions
The above descriptions of the limitations of original data and the normalisation approach to
benchmarking lead to the following summary cautions of which readers should be aware:
All original data and the benchmarking results derived from them are not 100%
comparable and have been graded to provide an indication of the
quality/comparability.
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Australia (40 degC) 1.778 1.752 1.726 1.842 1.973 1.789 1.790 1.484 1.442 1.479 1.389 1.285
EU (60 degC) 0.55 0.52 0.50 0.48 0.47 0.50 0.49 0.48 0.49 0.49
EU (43 degC) 0.88 0.84 0.80 0.78 0.76 0.80 0.79 0.78 0.79 0.79
0.0
0.5
1.0
1.5
2.0
2.5
N
o
r
m
a
i
l
s
e
d

s
a
l
e
s

w
e
i
g
h
t
e
d

c
o
n
s
u
m
p
t
i
o
n

(
k
W
h

i
n

a

s
t
a
n
d
a
r
d

w
a
s
h
)
Figure 23. Comparison of impact of varying test temperatures vs
consumer usage on normalisation results.
`
P a g e | 45 P a g e | 45



Normalisation has been based on the energy used for water heating only which has
limitations both in approach (and its particular sensitivity to variations in test
temperature and water consumption) and due to the optimisation of machines to local
requirements.
Individual performance attributes (wash quality, spin effectiveness, etc) of the
washing machines have not been normalised nor accounted for in the normalisation
of energy consumption.
conditions under test. The data and analysis on energy consumption and other
performance attributes presented in this report are based on reported testing
outcomes and are likely to be different from those experienced by the consumer.
Summary grading of mapping and benchmarking data
Based on the information on original data quality, the approach to normalisation of data for
benchmarking across countries and the associated cautions, a summary of gradings of both
mapping and benchmarking data and the associated summary rationale for each are
provided in Figure 24.

`
P a g e | 46 P a g e | 46



Figure 24. Summary classification of original data and benchmarking outputs.
Australia
Sales weighted source data is Robust, Product weighted source data is Indicative.
High quality dataset consisting of product level data based on mandatory national registration
system, sales weighted data more accurately represent actual products sold.

information is Illustrative. However, caution should be used when comparing with other
countries given the known consumer preference for cold wash (and hence significantly
lower energy consumption than that shown when normalisation is based on nominal test
temperatures.
Austria
All source data is Indicative.
Product and sales weighted source material based on aggregated market data (although original
source is product based with an estimated 90% coverage of the market).

Canada
Product weighted data based on high quality product level information. Sales weighted data
supplied as market averages with derivation of sales weighting unknown but believed to be
reliable. However, material based on Canadian standard measurement methodologies (e.g. unit
capacity based on drum capacity) which require conversions to equivalent comparable
measurements using established conversion methodologies, but which may not be accurate at
the individual product level.

Unit Energy Consumption data prior to 2004 is not 100% comparable with earlier data. In 2004
the Canadian and USA required test method/energy reporting algorithm was revised. This
revision included changes to the balance of wash temperatures, such that the hot and warm
wash cycles were assigned lower weighting factors after 2004. This revision, on average,
will tend to result in higher energy consumption values being reported prior to 2004 in
comparison with those after. It is also possible that manufacturers, anticipating the changes to
the test procedure which provided more credit to colder wash settings, incorporated additional,
cooler-temperature settings which would get averaged in to lower the rated energy use with no
physical changes to the machine. Drying energy was also included within the test which may
have resulted in manufacturers increasing spin speeds to give a net improvement in reported
efficiency, although the additional drying energy is not included in the data reported within this
report.
32

Benchmarking outputs: All information is Illustrative and comparisons with other
countries should be undertaken within the context of the cautions outlined above.

32
`
P a g e | 47 P a g e | 47



China
All source data is Illustrative.
Original source data supplied from the mandatory national registration system in the form of
averages. However, the protocol for expired products being removed from this database is
unknown (hence data may include products no longer available) and test procedure does not
provide a mandatory temperature requirement.

Benchmarking outputs: Due to the uncertain nature of the quality of this data source,
information on China is excluded from the analysis.
Denmark
Data believed to provide comprehensive picture of both models available and sales. However,
data presented based on each EU labelling category with specific model based energy
consumptions unknown and assumed to be mid-point of labelling category (which is likely to
result in slightly better energy performance than the true market average).

Benchmarking outputs: All information is illustrative.
Korea
Data based on compulsory product registration system (for top-loaders from 2001 and front-
loaders from 2007). However, significant assumptions were necessary for the date products
ceased to be on the market.

Benchmarking outputs: Data is Illustrative (front-loaders only). Due to the lack of
definition of a test temperature for top-loading units, these units have been excluded from
the benchmarking analysis.
Switzerland
Sales weighted source data is Indicative.
Only combined sales weighted data (i.e. both top and front-loader) available and supplied as
market averages. However, the original dataset from which averages are defined is believed to
be representative of the market as a whole.

Benchmarking outputs: Data is Illustrative (and is not sub-divided into top and front-
loading units)
UK
Front-loader sales weighted source data is Robust, front loader product weighted source
data is Indicative.
Data supplied on a model level basis and both product and sales weighted data believed to be
representative of the market. However, some issues with data reliability for the years 1999 to
2001.

Note the UK market is very strongly dominated by front-loading units with information on top-
loading machines very limited (in some years only 2 models are known to be available).
Therefore, to ensure no distortion of data analysis, benchmarking is limited to front-loading
machines only with this also believed to be representative of the market as a whole.

information is Illustrative.
`
P a g e | 48 P a g e | 48



USA
Sales weighted source material is Indicative.
Sales weighted data sourced from an industry body in aggregated form and considered to be
representative of the market. However, material based on USA standard measurement
methodologies (e.g. unit capacity based on drum capacity in gallons) which require conversions
to equivalent comparable measurements using established conversion methodologies, but given
data based on market averages rather than product level, conversions may not be completely
accurate.

Unit Energy Consumption data prior to 2004 is not 100% comparable with earlier data. In 2004
the Canadian and USA required test method/energy reporting algorithm was revised. This
revision included changes to the balance of wash temperatures, such that the hot and warm
wash cycles were assigned lower weighting factors after 2004. This revision, on average,
will tend to result in higher energy consumption values being reported prior to 2004 in
comparison with those after. It is also possible that manufacturers, anticipating the changes to
the test procedure which provided more credit to colder wash settings, incorporated additional,
cooler-temperature settings which would get averaged in to lower the rated energy use with no
physical changes to the machine. Drying energy was also included within the test which may
have resulted in manufacturers increasing spin speeds to give a net improvement in reported
efficiency, although the additional drying energy is not included in the data reported within this
report
33
.

Product weighted data is based on ENERGY STAR and is robust but represents only most
efficient proportion of the market.

Benchmarking outputs: All information is Illustrative and comparisons with other
countries should be undertaken within the context of the cautions outlined above.
EU
Data supplied on an aggregated market level. However, original data source is at model level
and is believed to be representative of the whole market.

Given the significant differences in measurement methodologies and reporting
protocols, all non-energy variables, i.e. load, water consumption, drying effectiveness
and wash quality, are presented as Illustrative.

33
`
P a g e | 49
Issue date: March 2012 Page 49
Benchmarking Document Washing Machines

The information and analysis contained within this summary document is developed to inform policy makers. Whilst the information analysed was supplied by representatives of National Governments, a number of assumptions, simplifications and transformations have
been made in order to present information that is easily understood by policy makers, and to enable comparisons with other countries. Therefore, information should only be used as guidance in general policy - it may not be sufficiently detailed or robust for use in setting
specific performance requirements. Details of information sources and assumptions, simplifications and transformations are contained within the document or the related Mapping Documents.

Annex 4 Policy summary table
The following table summarises the timetable for key mandatory and voluntary labelling requirements, minimum energy performance
standards (MEPs) and major voluntary industry actions in each of the countries/regions featured in the analysis.
Australia Canada EU Countries Korea Switzerland USA China
Pre-1990 Mandatory labelling
introduced in 1978
MEPs introduced in 1989
1990
1991
1992
1993
1994 MEPs introduced
1995 MEPS introduced
1996 Mandatory A-G Label for
efficiency (and wash and
spin performance).

Industry Voluntary
Agreement to remove E F
and G rated by 1999, and
D rated by 2003.

1997 Introduction of voluntary
ENERGY STAR labelling

1998 Introduction of Mandatory
Labelling of Energy
Consumption (6 Star
Rating System) including
minimum wash and spin
performance.

Requirement to register
product
Mandatory labelling
introduced

`
P a g e | 50


1999 Introduction of Voluntary
Eco-Label

2000 Voluntary certification
labelling introduced
2001 ENERGY STAR
introduced
Mandatory labelling.
Product registration and
MEPS for vertical axis
machines (1-5 scale)

2002 New A+ labelling category
adopted informally by
industry.

New Industry Voluntary
Agreement to improve
fleet average efficiency
by 2008 (to 02 kWh/kg)
Announcement of the
adoption of EU A-G
labelling

2003 Mandatory use of EU A-G
labelling

2004 Revised MEPs (with
associated revision to test
standard and required mix
of load/temperature tests)

ENERGY STAR
requirements revised
MEPs revised (with
associated revision to test
standard and required mix
of load/temperature tests)

ENERGY STAR

2005 MEPs and certification
level revised
2006 Inclusion of mandatory
rinse performance
Mandatory labelling,
product registration and
MEPS for horizontal axis
machines (1-5 scale)

`
P a g e | 51


2007 Inclusion of standby
power in derivation of
label rating.

Additional Water Label
introduced (also 6 Star
rating system)
ENERGY STAR
(including the addition of a
water requirement)

MEPS revised to
harmonise with the US
(no water efficiency
requirement in Canada)
Introduction of Maximum
Stand-by Power (off
mode) requirement.

Revised vertical axis
MEPS
MEPs and Water
Standard revised

Revised ENERGY STAR
product efficiency and
water requirements
Mandatory labelling (1-5
scale) and product
registration
2008 Addition of commercial
clothes washers to the
regulations

2009 ENERGY STAR
Revised MEPS for
horizontal axis machines
ENERGY STAR

2010
Post
2010
New ENERGY STAR
minimum energy and
water efficiency
requirement effective Jan
20, 2012

Anticipated new MEPS
standard in 2011 effective
2015 and 2018
Revised Energy Label
(A+++ to D)

MEPS revised :
1 December 2011
minimum requirements for
washing efficiency (class
A), energy efficiency
(class A) and water
consumption (according
to formula).

MEPS second revision:
1 December 2013
Revised Energy Label
(A+++ to D) with required
minimum A wash quality -
effective December 2011
New ENERGY STAR
product efficiency and
water requirements
(2011)

Anticipated new MEPS
standard in 2011 effective
2015 and 2018

Benchmarking WM

Transféré par

Informations du document

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Benchmarking WM

Transféré par

Droits d'auteur :

Formats disponibles

P a g e | 1 P a g e | 1

C wash. However, the

C with a cold water inlet of 15

C (recent revisions change wash

C wash) and top-loader units (water inlet temperature

C); and the same EU data using a nominal wash temperature of 43

C nominal temperature is used for EU

Vous aimerez peut-être aussi